Another Nikon D810A review and high ISO comparison

Another Nikon D810A review by Adam Woodworth with few ISO comparisons between the D810A, D810 and D750. Adam's conclusion:

In my experience the D810A matches the high ISO performance of the D750.  The D750 has a 24MP sensor and until the D810A came out it was the cleanest high megapixel DSLR, about a stop better than the D810.  But the D810A, like the D800/D800E/D810, has a 36MP sensor and appears to have the same high ISO performance of the 24MP D750.  With some other nice features, the D810A is an amazing landscape astrophotography camera although there are some gotchas caused but the IR filter.

The D810A is a landmark camera for astrophotography, with amazing high ISO performance and the convenience of (limited) exposures longer than 30s in camera. The IR Cut filter is both a plus and a minus depending on the situation, but the nice pop and increased color that it brings out in the Milky Way is very nice.

The Nikon D810A is currently in stock at Adorama, Amazon and Jessops (UK).

This entry was posted in Nikon D810a. Bookmark the permalink. Both comments and trackbacks are currently closed.
  • Spy Black

    “The D810A is a landmark camera for astrophotography…”

    This fellow does appear to have astrophotography background, but I wonder if astrophotographers in general share that enthusiasm for the D810A. They are typically spending (quite a bit) on astro-specific single-channel sensors for astrophography.

    • Eric Calabros

      After hearing many many different thoughts about what actually astrophotogs doing, because of D810A, now I’m officialy certain that I have no clue what those guys are doing and what equipment they prefer most.

      • Ric

        I bet they have beanies with propellers.

      • Judging from the stuff I read when googling the subject earlier, it seems like Canon has had a monopoly in this area for a long time (producing astronomy variants of its DSLRs for over ten years). So there are definitely some folks who want a better camera/sensor but don’t want to use a monochrome sensor.

        Sadly for Nikon, Sony just released the A7R II — which a lot of people will at least wait for before buying the 810A.

        Note that if you’re not doing very long exposures, you can use multiple filters with a monochrome sensor and build a color exposure (in fact, virtually all color pictures from NASA probes are made this way if they’re not flat out false color images).

        If you are doing long exposures with tracking, I suspect the DSLR’s colored filters make life a lot easier.

        • Wei-Hao Wang

          The reason Canon had a monopoly is that Nikon screwed up badly. The internal image processing routine in pre-2012 Nikon DSLRs could potentially remove stars, and this is hard to turn off (possible, but hard, google “mode 3”). This droves many astrophotographers away from Nikon.

          Now, it is found that Sony A7 series has this same internal image process that removes stars. (Try to find “A7 star eater” on dpreview.) So while Nikon is working hard to win back astrophotographers, Sony is repeating exactly the same mistake Nikon had made 10 years ago. I found this ironic, and disappointed too as I just bought a used A7R.

    • neonspark

      uh, who so what? It is a product for a nice audience. if you’re a niche within a niche and this isn’t for you, does it matter? Clearly canon and now nikon thought a simple mod to an existing camera was worth it in order to sell more gear with little investment on their part.
      You may look at this and say: aha! true astro-photogs shooting with cooled CCDs hate it! But whatever. Nikon never set out to kill the hubble space telescope either. This is just a product for a type of astro-photog. Contrary to your assertion, this is not a clearly defined category. Just as with ALL types of photographers, people then go into sub-niche areas.
      Overall, I see this as more choice, which is good considering that should I ever be so inclined to shoot stars and nebula without buying time on top of a volcano in Hawaii, this camera is an option I didn’t have before.

      • Spy Black

        I wouldn’t know, that’s why I asked the question.

      • TheInfinityPoint

        Lol, I get time on those telescopes and take photos of them.

    • RIT

      I guess “landscape astrophotography” is the key difference between dedicated astrophotography sensors and the D810a. I for one would certainly appreciate less noise and more colour in night landscapes with a DSLR which a dedicated sensing apparatus might be less suitable for. But would a D810a with minimal post processing beat a firmware moded D800e with a little more work in post?

      • Dave_D69

        Yes it would…

        • Spy Black

          Why is that (inasmuch as I’ve never heard of a modded D810)?

          • ITN

            4x sensitivity in a certain wavelength range => 2x SNR if limited by photon noise, up to almost 4x SNR if limited by read noise. The improved SNR cannot be recreated in a firmware modification. Of course if you modify the optical filtration on the sensor and write new firmware which would give the night live view performance of the D810A then you could do it but how much time would that take to reverse engineer the firmware and develop the new code vs. just buying a factory made D810A. I think whenever a product that fits a purpose exists and is commercially available it is cheaper to buy it than do the development work yourself.

            • Wei-Hao Wang

              The firmware hack of D800/D800E only restores the true raw data in the NEF files. This substantially improves the calibration (dark subtraction and flat fielding) for astrophotography. In some sense, because of the better subtracted dark, this does improve the image S/N in critical cases. However, in general, especially in casual astrophotographers’ hands, this has almost no impact to image quality.

              On the other hand, as ITN mentioned, the 4x more H-alpha sensitivity in D810A is a deal breaker for almost everyone who points the camera on the deep sky. The same sensitivity can be achieved on D800/D800E/D810 with a hardware modification. Such a modification is a default for serious astrophotographers for more than 10 years, except for those who own Canon’s 20Da, 60Da, and now Nikon’s D810A. At this moment, the best C/P values on D8x0 series for astrophoto would be a used D800 plus hardware and firmware modification. However, D810A does give you the same high H-alpha sensitivity out of box without any modification, and some other nice (but not super critical) features such as M* and internal intervalometer, etc. You get what you pay for.

          • Dave_D69

            Exactly. It beats the unmodded one.. the difference isn’t in the firmware.. rather the sensor.. the D810 and D810A sensor is simply better than the D800..

            • Wei-Hao Wang

              May I ask where I can find information or test results that show the sensor of D810 is better than the sensor of D800? Based on DxOMark numbers, there is almost no apparent difference between D810 and D800.

      • Spy Black

        I’ve never heard of a firmware modded D800 of any incarnation.

  • Purdyd

    If Nikon was truly serious about astrophotraphy they would have added a 1:1 video mode that could be windowed to as little as 1/4 vga or 4k and could output high fps – in a raw format

    The could also produce a high quality f mount to t thread adapter

    The should also work with some of the popular freeware software package authors of astrophotographers software to get the camera integrated

    • Patrick O’Connor

      Can you explain, or point to an explanation of, what all that means relative to astrophotograpy?

      • Purdyd

        If your photographing a planet, they are quite small and the limitation is reach and the atmosphere

        So the solutiion is to use small pixel pitch sensors and take lots of images, select the best , and stack them.

        Over a period of several seconds there will be random times when the planet is much clearer

        You stack the best pictures and stacking decreases the noise and also increases the resolution

        If you do deep sky wide stuff the dslr comes into its own but those images are also stacked reduce blur caused in telescope mount tracking

        There is a variety of software to automate the capture and stacking of images, you will find canon cameras supported much more than Nikon

        If you are hook in up a dslr you will need an f mount to either s t thread or 2″ tube and there is no real quality adapter

        Google and hang out in some of the astronomy forums and you can learn quite a bit in a short time

        I wonder why Nikon didn’t do the same.

        • If you want to do this kind of thing with the D810A (or other Nikon DSLRs) a better option would be tethering the camera to a computer which could do all kinds of magic (including allowing you to flag which images to stack in real time).

          Removing the filter from a sensor is easy and doesn’t second-guess how the camera will be used. Investing time and money developing features unique to the D810A that would (a) delay its release, (b) probably not be exactly what any actual user wanted wouldn’t make sense.

          • Purdyd

            Nothing I suggested can be done externally to the camera.

            There is no reason to shoot a 36mpix image of a planet with a full frame sensor

            A popular feature for canon cameras is back yard eos, look it up.

            All of the features could have been tested well before the sensor was changed.

            Having the Nikon camera raw files be compatible with more astrophotography software was exactly my point.

      • Nimloth

        Video = stacking for higher resolution images of bright objects (i.e. planets). It’s exactly the same principle as the sensor shifting mirrorless cameras use, except with video, you can stack hundreds or thousands of frames. Planets rotate, so you want as many frames possible in a short time – hence high fps.

        F to T adapter is for attaching the D810a to a telescope or other optical system, many of which use this thread type. 🙂

        Software collaboration would be the intelligent thing to do. You cannot make the intended astro images without software, which means not only PS/LS, but more or less obscure packages to do odd things. Many of the popular ones are freeware, and it’s likely these ones the intended market for this low price camera would use.

  • Nimloth

    I would love to see Wei-Hao Wang give the D810a a go, and make a review/guest post about it here. He had some excellent explanations of using the D810 for astro work, and wrote a really nice astro tutorioal on photographylife a while ago, probably other stuff I can’t think of right now too. 🙂

    • Wei-Hao Wang


      Surprised to see my name mentioned here. Unfortunately I don’t have a D810A. I have two D800s modified for astrophotography, so I don’t have an urgent need for a D810A.

      On the other hand, a few friends of mine got their D810As. One of them is kind enough to shoot many test exposures and provide them to me for analyses. I was able to measure its readout noise and dark noise and compare the results with my D800. Not surprisingly, D810A’s dark noise is substantially lower than D800’s.

      On the other hand, it surprises me a lot that D810A’s readout noise is 30% to 50% higher than D800’s, depending on ISO. (When comparing D810 and D800’s readout noise on, which are based on DxOMark, I also found a similar trend: D810 appears to be a bit more noisy in readout noise than D800.) This is quite puzzling. In normal daytime high-ISO shots, readout noise is the dominant noise term. If D810A is noisier than D800, how come several people found its high-ISO performance better than D800? One possibility is a higher quantum efficiency, which can compensate the larger readout noise. This is possible since the filter in front of D810A’s CMOS allows more light to go through. The fact that D810A’s base ISO is 200 also confirms that its sensor sees a higher photon flux than normal D800 and D810. However, I am not sure whether just this factor alone can explain everything we saw.

      I have no doubt that D810A will be a great camera for deep-sky astrophotography and astro-landscape, better than my beloved D800-mod. I am sure many exciting pictures will come out as time moves on. I am personally more interested in the technical aspect of D810A, such as if there is any real difference between the readout system of D810A, D810, and D800, and exactly what is its quantum efficiency, etc. Without having my own D810A it will be hard to conduct thorough tests. Perhaps later when DxOMark and sensorgen release their numbers on D810A we will see a clearer picture.

      • Nimloth

        Hi. This is exactly why I mentioned you: In this one post you bring some good and useful information and clarification not mentioned elsewhere (on regular photog sites anyway). Also a bit of tech geekiness (and geeks are awesome people) which is probably exactly the thing those who seriously consider this camera need to know. Most importantly, it’s based on experince since you actually use a similar system regularlyfor serious astro stuff; it’s not something you only test. Love your photos, BTW. 🙂

  • KT

    Sony Just released their much anticipated 42 MP BSI sensor, time to move on from the 36 MP

      • mrterrabyte

        Peter, haven’t Nikon (when using a new release and to-be-shared Sony sensor) usually announced their new model first, e.g. D3x, D8**?

        Perhaps Nikon’s waiting for Canon’s 50mp 5Ds (launching here early next week in Aus) to falter first…

        • That was the case in the past, but maybe now Sony no longer let them go first and want to get all the attention first – it’s their sensor after all, why let anyone else premier it.

          • mrterrabyte

            Guess we’ll see! thx, hadn’t intended to hijack the thread lol Good article and review by the author btw

  • Back to top